Malignant gliomas (MGs) remain one of the most deadly of all cancers affecting Veterans and are incurable. It is clear that different approaches are needed to better study the basic and translational science of these tumors. One of the hallmarks of MGs and most cancers are the differences among the cancer cells from the same patient with regards to ability to proliferate and form tumors in vitro and in vivo. This has led to the hypothesis that a subpopulation of cancer cells has the properties of brain tumor-initiating cell or cancer stem cells (CSCs), which are highly enriched for the ability to form tumors following transplantation as compared to bulk tumor cells or non-tumorigenic tumor cells. CSCs have been demonstrated in a number of cancers such as human leukemia, breast cancer, and colon cancer by identifying a cell surface marker that distinguish CSCs from cancer cells with more limited proliferative potential. However, the cell surface phenotype needed to identify CSCs in MGs remains unclear. Moreover, it is unlikely that only one marker will suffice and it may be the case that combinatorial markers will be required to truly identify CSCs due to the complex heterogeneity of many cancers and the possibility of more than one CSC subpopulation, particularly when there may be different subtypes such as seen with MGs. Because there are a relatively large number of cell surface markers that could possibly identify CSCs in MGs and other cancers, there is a critical need to develop a powerful screening technique or approach to accurately identify one or several candidate markers before proceeding with confirmatory testing using the time-consuming and costly in vivo mouse xenograft assays. Currently, there is no such approach that can screen hundreds of cell surface markers on living tumor cells to identify CSC sub-populations within MGs or any other human cancer. We propose to establish such a strategy by developing a novel live cell microarray, which contains thousands of printed antibody microspots, each one with the potential to capture a specific subpopulation of living cancer cells based on the cell surface marker expression for each cell. It is then determined which of the cells spot(s) have the capacity to form cancer colonies on the array when grown in CSC conditions. We can screen hundreds or thousands of cell surface markers on living cancer cells and identify those that enrich (and do not enrich) for CSC behavior on the array itself in vitro, and thus, serve as candidate positive (and negative) CSC markers. This novel technique, called the Colony-Forming Antibody Cell Array (CFACA), has been initially developed using traditional MG cell lines and late-passage MG CSC lines to establish specificity and sensitivity of cell capture, and capacity of colony formation. We now propose to optimize the CFACA to identify CSCs from the critical MG cells of freshly dissociated surgical specimens or after early passage in culture. The CFACA-derived candidate CSC markers will be assessed with various in vitro studies and mouse xenograft assays to validate the CFACA as a novel screening assay to identify the CSC cell surface phenotype (or phenotypes) in MGs. The CFACA will also obtain the cell surface marker expression profile of the MG CSCs in relation to other cells within the same tumors, and the CSC phenotype(s) will be correlated with clinically relevant aspects of MGs such as histology subtypes, methylation of MGMT promoter, and overall patient survival. The identification of CSC subpopulation(s) in MGs will provide insights into pathogenesis and establish previously unidentified cellular target(s) for more effective diagnostics and therapies of MGs. We believe that the CFACA has the potential to one day serve as a patient-specific diagnostic tool for MGs, by identifying the unique CSC subpopulation for each Veteran patient, which in turn, will help direct a more specific therapy regimen. While the focus of this proposal is to study MGs, the CFACA has the potential to identify or refine the CSC phenotype(s) in any human cancer affecting Veterans, and even identify somatic stem cells in a variety of normal human tissues.